Видео

I love you, you make my life easier 🥰

Great job! Thanks!

Thank you for the video! I have tried this and works fine on desktop but the CascadeClassifier returns null on Android. Could :c

thank you so much! Hope you make more tutorial opencv

I have a problem in outputArrayOfStockList.cs. How to solve it?

Amazing tutorial but this detecting other things of my room sometimes my face how i fix this any suggestions???

Does anyone have the archive for the elements from Flappy Bird?

I have this problem. I searched for so many ways to tackle it but always failed. DllNotFoundException: OpenCvSharpExtern assembly:<unknown assembly> type:<unknown type> member:(null) OpenCvSharp.NativeMethods.TryPInvoke () (at Assets/OpenCV+Unity/Assets/Scripts/OpenCvSharp/PInvoke/NativeMethods.cs:115) Rethrow as OpenCvSharpException: OpenCvSharpExtern assembly:<unknown assembly> type:<unknown type> member:(null) *** An exception has occurred because of P/Invoke. *** Please check the following: (1) OpenCV's DLL files exist in the same directory as the executable file. (2) Visual C++ Redistributable Package has been installed. (3) The target platform(x86/x64) of OpenCV's DLL files and OpenCvSharp is the same as your project's. System.DllNotFoundException: OpenCvSharpExtern assembly:<unknown assembly> type:<unknown type> member:(null) OpenCvSharp.NativeMethods.TryPInvoke () (at Assets/OpenCV+Unity/Assets/Scripts/OpenCvSharp/PInvoke/NativeMethods.cs:115) OpenCvSharp.NativeMethods.TryPInvoke () (at Assets/OpenCV+Unity/Assets/Scripts/OpenCvSharp/PInvoke/NativeMethods.cs:124) OpenCvSharp.NativeMethods..cctor () (at Assets/OpenCV+Unity/Assets/Scripts/OpenCvSharp/PInvoke/NativeMethods.cs:59) Rethrow as TypeInitializationException: The type initializer for 'OpenCvSharp.NativeMethods' threw an exception. OpenCvSharp.CascadeClassifier..ctor (System.String fileName) (at Assets/OpenCV+Unity/Assets/Scripts/OpenCvSharp/modules/objdetect/CascadeClassifier.cs:39) NewBehaviourScript.Start () (at Assets/Scripts/NewBehaviourScript.cs:16)

Anyone know how to release the camera like in normal opencv with cap.release?

Excuse me, do you have github address? I want to refer to it for my class assignment

Can I move it with markers?

I am having this error " NullReferenceException: Object reference not set to an instance of an object FaceDetector.findNewFace (OpenCvSharp.Mat frame) (at Assets/FaceDetector.cs:32) FaceDetector.Update () (at Assets/FaceDetector.cs:26)" ... my plane material is not switching to instance ..please help

I have a problem the video through the camera is upside down. Please help me to rectify this

Hi! I´m having a problem, could anyone help me? I followed the steps, but I can´t use the faceY variable, the way of calling it doesn´t seem to work for me.

It tells me: FileNotFoundException: "/Users/***/***/***/MotionController3/Assets/haarcascade_frontalface_default.xml"not found OpenCvSharp.CascadeClassifier..ctor (System.String fileName) (at Assets/OpenCV+Unity/Assets/Scripts/OpenCvSharp/modules/objdetect/CascadeClassifier.cs:40) FaceDector.Start () (at Assets/Scripts/FaceDector.cs:18) It's exactly as in the video.

How to get last frame value??

a easy way to fix the Null error with the cascade is to manually download the Haarcascade file (its on geeks for geeks) and simply drop it on the assets folder in your project. Thats where his path is looking for it. Everything else seems to work fine

笑死。看这个tutorial的时候刚好是海康威视在美被禁

Has anyone successfully run this face recognition feature on Android? I installed it directly on the phone and found that even the camera function does not work properly 😓

can not detect on android device

Nice! Thank you!

Amazing, thank you! I could use some help with landmarks and don't know where else to ask. Maybe someone has an idea? OpenCV plus made me download a shape predictor, of which I changed the file name from .dat to .bytes, I've seen it work this way in the demo scene. The idea is to track the mouth using point 63 out of 68 so I can later check if the mouth has opened and then do stuff. Code: private ShapePredictor predictor; public TextAsset shapes; //the downloaded LandmarkLib private Point mouth; private void Start() { byte[] shapeDat = shapes.bytes; predictor = new ShapePredictor(); predictor.LoadData(shapeDat); //Telling the shape predictor to use the downloaded library } void Update() { myFace = //find the face like in this Tutorial, all working findLandmarks(frame, myFace) } void findLandmarks(Mat frame, Rect face) //this makes Unity crash { if (predictor.DetectLandmarks(frame, face)[63] != null) { mouth.X = predictor.DetectLandmarks(frame, face)[63].X; mouth.Y = predictor.DetectLandmarks(frame, face)[63].Y; } } How else am I supposed to track a point on the face without Unity crashing & saying "This application has requested the runtime to terminate it in an unusual way(...)"? Is it crashing because it is running slowly? Have I referenced or imported the shape_predictor_68_face_landmarks.bytes in the wrong way? There surely must be a way to use ShapePredictor.DetectLandmarks, right? Any help is appreciated <3

Thanks for a great tutorial - I had to add the "lastY=faceDector.faceY;" line suggested by someone below, unsure if it's supposed to work without this, but it did the trick for me. However, the face tracking is really jittery and imprecise for me, it jumps between different values even when my face is completely still. Does anyone know a good way to calibrate the tracking, or alternatively, to smooth out the transition between values?

Everything is working fine, but my bird in game is still flying upwards despite the updating the code as shown in the video, can anyone help me on this? what could possibly the issue? Below is the code for the bird: using System.Collections; using System.Collections.Generic; using UnityEngine; public class BirdController : MonoBehaviour { FaceDector faceDector; float speed = 1f; float lastY = 0; public void Start() { faceDector = (FaceDector)FindObjectOfType(typeof(FaceDector)); } public void Update() { float step = speed * Time.deltaTime; float norm = Mathf.Clamp(faceDector.faceY - lastY, -1, 1); transform.position = Vector3.MoveTowards(transform.position, new Vector3(transform.position.x, transform.position.y + norm, transform.position.z), step); } }

Thank you for your video! it was very interested, and I have a question, Is possible to deploy a project in WebGL instead of android or ios?

Also if you build the game/program: place the haarcascade_frontalface_default.xml directly in the first folder in unity (assets) and change the Application.dataPath to System.IO.Directory.GetCurrentDirectory + @"/path where the xml file is in the build folder" The build cant find dataPath and also not when its in a under directory.

If somebody has problem with the haarcascade_frontalface_default.xml. In my case it wouldnt found this file till i added a / infront of the path: cascade = new CascadeClassifier(Application.dataPath + @"/haarcascade_frontalface_default.xml");

hi very amazing thid toturial is very useful........................... pleas continue learn open cv from basic in windows and android plarform

Hi, Thank you so much for the great tutorial. I am going to use OpenCV for my Unity project. In that project I need to implement Jump, Run, SwipeLeft, SwipeRight and etc gestures with OpenCV using a Web camera. Can you please confirm that the Jump, Run, SwipeLeft, SwipeRight, and etc gestures are possible with OpenCV pose tracking? If possible can you please help me or share some of the references for the above gestures implementation using Unity. Thanks in advance.

Can u do the same with hand detection instead of face detection using OpenCV+Unity?

Hi, Can you please share a few references for the skeleton tracking detection using OpenCV in Unity.

Awesome can you create more tutorial

i have a problem. How to fix haarcascade_frontalface_default.xml"not found.

Awesome tutorial

Hi, very good tutorial. Can Open CV works on integrated graphics like intel hd4000?

Thank you fro the awesome tutorial! Do you know where to find any documentation on this? the domain in the asset store seems to not exist anymore. I was trying to build off this tutorial to detect facial landmarks by making another cascade with the shape_predictor_68_face_landmarks.bytes file, but the file seems to be empty so it doesn't work the same way. Do you know how to find facial landmarks using this asset? I can't seem to find anything about it anywhere.

ok)

Absolutely great video! Thanks very much

Nice! :)

Espetacular, Seria muito bom também com o jogo tetris usando detecção de mão

I get a NullReferenceException error in the findNewFace function, how could I resolve this

Nice! :)

Where did you get the assets you used in this tutorial? I searched flappy bird on the unity asset store and these materials don't come up.

Hi , Great tutorial , but is there a way to get the rotation of the face ???

I got stuck on the part where the bird keeps flying of screen instead of following the rectangle. This is the code from the facedetector.cs: using System.Collections; using System.Collections.Generic; using UnityEngine; using OpenCvSharp; public class FaceDetector : MonoBehaviour { WebCamTexture webCamTexture; CascadeClassifier cascade; Renderer Render; public float faceY; OpenCvSharp.Rect MyFace; // Start is called before the first frame update void Start() { //Gets the renderer component and stores it in the variable Render Render = GetComponent<Renderer>(); //Stores all the webcam devices it has found into the array devices WebCamDevice[] devices = WebCamTexture.devices; //Here we set the webcam texture to the first webcam device that has been found webCamTexture = new WebCamTexture(devices[0].name); //Then we play the webcam input webCamTexture.Play(); //Here we create a new cascade classifier from the dataPath provided by the program cascade = new CascadeClassifier(Application.dataPath+ @"/Haarcascade/haarcascade_frontalface_default.xml"); } // Update is called once per frame void Update() { Mat frame = OpenCvSharp.Unity.TextureToMat(webCamTexture); findNewFace(frame); display(frame); } void findNewFace(Mat Frame) { //Gets the face that currently in the frame and if it matches, stores it in the variable faces var faces = cascade.DetectMultiScale(Frame, 1.1, 2, HaarDetectionType.ScaleImage); if(faces.Length >= 1) { Debug.Log(faces[0].Location); MyFace = faces[0]; faceY = faces[0].Y; } } void display(Mat frame) { //First we check if MyFace isn't empty. If so we add a rectangle on the face if(MyFace != null) { frame.Rectangle(MyFace, new Scalar(250, 0, 0), 2); } //Here we add the face as a texture to the variable newTexture Texture newTexture = OpenCvSharp.Unity.MatToTexture(frame); //Then we set the newTexture as the main texture on the renderer Render.material.mainTexture = newTexture; } } Down here is the code from the bird.cs: using System.Collections; using System.Collections.Generic; using UnityEngine; public class Bird : MonoBehaviour { public FaceDetector faceDetector; float speed = 5f; float lastY = 0; private void Start() { faceDetector = (FaceDetector)FindObjectOfType(typeof(FaceDetector)); } // Update is called once per frame void Update() { //Moves the bird overtime with a set speed float step = speed * Time.deltaTime; //Stores the value of the last Y position in the previous frame and then calculate the difference between them float norm = Mathf.Clamp(faceDetector.faceY - lastY, -1, 1); //Afterwards we normalize the difference transform.position = Vector3.MoveTowards(transform.position, new Vector3(transform.position.x, transform.position.y + norm, transform.position.z), step); } } If someone could tell me what I did wrong please let me know!!

Please provide object detection by open cv

very good tutorial . anyway is that possible for test it without webcam on pc? just wanna try directly build on the smartphone? thanks in advance.

thats great tutorial, but it really slow, only 10fps when run (My pc i7 6700 + Nvidia 1060 ). How to improve performance?

A quick tutorial for a simple yet awesome application using only free tools. And you even took your time for some reflections at the end. Good stuff!

thank you very muchhhh🤩🤩🤩🤩🤩